If you are asked to discuss code in an academic integrity review, focus on how you developed it and whether you can explain its behavior.
A detector score alone cannot establish who wrote code or whether a rule was broken, so understanding the evidence and your institution’s process matters.
Understand what the review is examining
Start by clarifying the concern.
An instructor or review panel may be asking about unauthorized assistance, copied code, collaboration, use of generative AI, or a mismatch between submitted work and your demonstrated understanding.
These are related issues, but they are not interchangeable.
Ask which assignment, files, and course rules are involved, and what evidence you are invited to address.
Distinguish code similarity from AI authorship classification.
Similarity systems compare code passages or structures against other submissions and known sources.
A match may point to material that needs explanation, but common algorithms, starter code, required libraries, and standard patterns can appear in many projects.
A similarity result is a lead for review, not a complete finding of plagiarism.
AI detectors attempt to classify whether content may have been generated by a model.
That kind of score is not proof of authorship or misconduct.
Turnitin has stated that its AI writing model does not reliably detect code; its AI writing report should not be treated as a source-code similarity report.
The tools answer different questions.
Review the course rules before responding
Read the assignment instructions, syllabus, and any applicable academic integrity guidance.
Look for rules on generative AI, code assistants, collaboration, tutoring, reuse of previous work, external libraries, and citation.
Policies vary among institutions and courses, and permitted use may depend on the assignment.
Do not assume that a general campus rule answers every question about a specific task.
If the rule or allegation is unclear, ask for clarification through the process described in the notice.
You can ask what conduct is under review, what materials are being considered, the response deadline, and whether you may bring an adviser or support person.
Follow the stated procedure, and keep copies of relevant messages and documents.
Gather records that show your development process
Collect authentic records that already exist.
Useful materials may include version-control history, dated backups, editor history, development notes, test output, assignment drafts, and messages about permitted collaboration.
Keep original files where possible.
A repository’s commit history can help show how a project changed over time, but it does not by itself prove who made every change.
Make a simple timeline with dates and specific actions: when you began, what you implemented, when you tested or revised it, and what assistance or tools you used.
Be precise about uncertainty.
Do not create retrospective records, alter timestamps, or present a reconstructed timeline as a contemporaneous log.
- Preserve the submitted version and any relevant earlier versions.
- Keep assignment instructions and the policy version that applied to the work.
- Record which libraries, starter files, examples, or collaborators contributed to the project.
- Save relevant tool-use records if they exist and the policy permits their use.
Prepare to explain what the code does
Read through your submission and be ready to describe its structure in plain language.
Explain the program’s inputs, outputs, main functions, data flow, and important design choices.
For a particular flagged section, discuss its purpose and how it connects to the rest of the program.
If you cannot explain a part, say so honestly rather than guessing.
Practice tracing a representative input through the program.
Be prepared to explain how conditions, loops, data structures, error handling, and tests work where relevant.
You do not need to claim perfect recall: code may have been written over several sessions, and developers use documentation.
The goal is to demonstrate understanding and give a credible account of your work.
If a similarity report identifies a passage, compare the exact code and surrounding context.
Explain any shared starter code, required implementation, common library usage, or source you consulted.
If you used a source or received help, describe that accurately and explain whether and how it was acknowledged.
Avoid dismissing a match without reviewing it.
Give a clear, factual response
Organize your response around the specific concern, the relevant course rule, and evidence that helps explain what happened.
Separate what you know from what you infer.
A calm, direct account is easier to assess than broad claims about what a detector can or cannot do.
You might structure a written explanation like this:
- Identify the work: name the assignment and the section being discussed.
- Describe your process: summarize when and how you developed the code.
- Address the evidence: explain the flagged passage or score in context, without treating it as conclusive.
- Disclose assistance: state relevant collaboration, sources, or tools accurately.
- Attach records: point to dated materials that support your account.
Do not delete files, coordinate stories with classmates, or alter records once a concern has been raised.
If you made a mistake, explain it plainly.
If you disagree with an interpretation, identify the specific point and the evidence that supports your view.
Understand the limits of detector results
Automated results can be affected by the material being compared, the tool’s design, and the context of the assignment.
A source-code similarity match can identify overlap without explaining why it exists; an AI authorship score cannot independently verify the author.
Neither result substitutes for reviewing the code, relevant rules, and your explanation.
Codespy is an external source-code detection service.
If it is mentioned in a review, ask what material was submitted to it, what comparison or output is being relied upon, and how that output was interpreted.
Do not describe it as having proven authorship or misconduct.
This article does not report independent testing or endorse a particular detector.
Ask questions and keep the process organized
Track deadlines and respond using the channel specified in the notice.
Keep your explanation focused, retain copies of what you submit, and ask how to provide additional evidence if needed.
If the possible consequences are serious or you are unsure about your rights under institutional procedures, consult an appropriate student advocate, adviser, or qualified professional.